Aircraft sensor data synchronous acquisition method based on interpolation method

Abstract

The invention discloses an aircraft sensor data synchronous acquisition method based on an interpolation method, and the method comprises the steps: carrying out the asynchronous sampling of each path of aircraft sensor on the premise of the same sampling frequency; and receiving each path of sampling value and screening to obtain a credible sampling value, and calculating the sampling time of each path of sampling value according to the transmission delay. A fixed time sequence is selected, a correction value corresponding to the sampling value is calculated through an interpolation function and serves as a new sampling value, the sampling time is the same fixed sequence, and data synchronization of the sampling value is achieved. Under the condition of ensuring the same sampling frequency, a synchronous signal is not needed to control the sampling time, and a corresponding synchronous signal hardware circuit is not needed, so that the production cost is remarkably reduced, the measurement precision of the sensor is remarkably improved, and relatively good practicability is achieved.

Description

Technical Field

[0001] The invention belongs to the technical field of liquid level measurement, and in particular relates to an aircraft sensor data synchronization acquisition method based on an interpolation method. Background Art

[0002] At present, the fuel measurement system based on computer technology needs to detect, correlate, combine and estimate the multi-source data provided by the fuel quantity sensor to achieve the purpose of improving the comprehensive utilization of information and the effectiveness of the entire fuel system. This is of great significance to improving the accuracy of aircraft status assessment and the precision of fuel quantity measurement.

[0003] However, when fusing multiple aircraft sensor data, a single data source is no longer sufficient for predictive analysis. In practice, data fusion methods are rarely applied throughout the entire data processing process, and they lack fault tolerance. If a single sensor erroneously outputs fuel level data, the resulting processing may be abnormal, or even significantly different from the actual fuel level, seriously impacting the pilot's ability to accurately assess the aircraft's operating status.

[0004] In summary, using accurate data fusion methods to perform multi-time and multi-angle detection on sensor sampling data, expanding the temporal and spatial coverage of sensor detection, reducing information uncertainty, enhancing system accuracy, making the system more anti-interference, and improving the ability to maintain continuous operation and weaken faults, plays an important role in system reliability. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for synchronously collecting aircraft sensor data based on interpolation method, aiming to solve the above-mentioned problems.

[0006] The present invention is mainly achieved through the following technical solutions:

[0007] A method for synchronously collecting aircraft sensor data based on an interpolation method comprises the following steps:

[0008] Step S1: Under the premise of the same sampling frequency, each aircraft sensor performs asynchronous sampling;

[0009] Step S2: Receive each sampling value and filter to obtain the credible sampling value U={u0,...,u i ,...,u n-1}, obtain the initial sequence of sampling values; where i = 0, 1, ..., n-1, and n is the number of samples of the aircraft sensor; then, calculate the sampling time of each sampling value based on the transmission delay;

[0010] Step S3: using the sampling value and the sampling time as two columns of original data, an interpolation function of each sampling value is obtained based on an interpolation algorithm;

[0011] Step S4: Select a fixed time sequence, calculate the correction value corresponding to the sampling value by the interpolation function of step S3, and use it as the new sampling value. The sampling time is fixed in the same sequence to achieve data synchronization of the sampling value.

[0012] In order to better implement the present invention, further, step S2 includes the following steps:

[0013] Step S21: Set the filter queue address, confidence interval length, and filter limit Δ limit And the upper limit of the filter queue Δ upper and the lower limit Δ lower ;

[0014] Step S22: Using the filter limit Δ limit Keep out obviously false data;

[0015] Step S23: Using the upper limit Δ of the filter queue upper and the lower limit Δ lower Filter the sample values and get the values that fall within [Δ lower , Δ upper ] interval of credible sampling value;

[0016] Step S24: Divide the original sampling values detected by the aircraft sensor into n units according to the original order to determine the initial sampling value sequence.

[0017] In order to better implement the present invention, further, in step S2, the transmission delay is determined according to the time from the sensor collecting the signal to the filter processing of the sampling value; the transmission delay is a constant or a measurable variable.

[0018] In order to better implement the present invention, further, step S3 includes the following steps:

[0019] Step S31: First, calculate the difference quotient. The interpolation quotient of each order of the sample value is expressed as:

[0020] First-order difference quotient expression:

[0021]

[0022] Second-order difference quotient expression:

[0023]

[0024] By analogy, the expression of the N-order difference quotient is:

[0025]

[0026] Step S32: Then, establish the interpolation function P of each sampling value n (u) is:

[0027] P n (u)=f(u0)+f[u0,u1](u-u0)+f[u0,u1,u2](u-u0)(u-u1)+…+f[u0,u1,…,u n ](u-u0)(u-u1)…(uu n-1 ) (8)

[0028] Where u is the sampling correction value at the current moment.

[0029] In order to better implement the present invention, further, in the step S4, when obtaining three consecutive discrete moments t0, t1, t2 and the sensor sampling values at the corresponding moments [t0, u(t0)], [t1, u(t1)], [t2, u(t2)], the interpolation function P is used n (u) The correction result of the sampling value u(t) at time t is:

[0030]

[0031] In the case of equal sampling intervals, formula (9) is simplified to:

[0032]

[0033] Where T is the sampling interval.

[0034] In order to better implement the present invention, further, the interpolation function error R2(t) is:

[0035]

[0036] Wherein, u′″(ξ) is the third-order derivative of the sample value u(t) at time t at t = ξ, ξ∈[t0,t2].

[0037] In order to better implement the present invention, further, in step S4, for each measured sample value u i (t i ), set d adjacent measurement points as a filtering reference interval, and the filtering reference interval moves m measurement points one by one over time to update the iterative filtering sequence; then use formula (10) to iterate the result of the corrected sampling value queue and enter the final output filtering queue.

[0038] In order to better implement the present invention, further, if it is known that the permissible trust threshold increment between adjacent sample values is Δ, then the method for eliminating p-fold errors of each sample value is:

[0039] if |ui (t i )-u i-p (t i-p )|≥p×Δthen u i (t i ) is corrected using the following formula:

[0040]

[0041] Where: j is greater than The smallest integer;

[0042] u' i (t i ) is the corrected result.

[0043] In order to better realize the present invention, it is further applied to an aircraft fuel quantity sensor.

[0044] The beneficial effects of the present invention are as follows:

[0045] While ensuring consistent sampling frequency, this invention eliminates the need for synchronization signals to control sampling timing and corresponding synchronization signal hardware circuitry, significantly reducing production costs. It also enables trend estimation of aircraft fuel quantity measurement data, bringing it closer to the actual state and eliminating fuel level fluctuation errors. Through error discovery, the invention allows for credible threshold increments and shifts the filter reference interval to update iterative Newton interpolation results, significantly improving fuel measurement accuracy, offering superior performance, and improving practicality. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a flow chart of the method for synchronously collecting aircraft sensor data based on the interpolation method of the present invention;

[0047] Figure 2 This is the principle diagram of the filter reference interval movement. DETAILED DESCRIPTION

[0048] Example 1:

[0049] A method for synchronously collecting aircraft sensor data based on interpolation method, wherein the sampled data is processed based on a filter, and the filter processing includes the following steps:

[0050] (1) First, a filter is set to post-process the sensor detection signal. In the filter, the filter queue address, the sampling value queue length and the upper limit of the filter queue Δ are set. upper and the lower limit Δ lower ;

[0051] (2) Then, using Δ limit Reject obvious false data; then use the upper limit Δupper and the lower limit Δ lower To judge the sampling value; to use Δ limit Reject obvious false data and use the upper limit Δ upper and the lower limit Δ lower The sampling values are judged and the reliability relationship between the sensor sampling values is analyzed to obtain the value falling within [Δ lower , Δ upper ] interval of credible sampling values.

[0052] (3) Then, the fuel quantity measurement sampling values detected by the sensor are divided into n units according to the original order to determine the initial sampling value sequence.

[0053] (4) The transmission delay is determined based on the time from when the sample value is collected by the sensor to when it is processed by the filter.

[0054] (5) After calculating the difference quotients of each order of the sampling value, the interpolation function P can be established n (u) expression.

[0055] (6) Finally, according to the interpolation function P n (u) Obtain the correction result of the sensor sampling value at a fixed moment, so as to optimize the performance of the filtering algorithm.

[0056] Preferably, it falls within [Δ lower , Δ upper The credible sampling value of the interval is:

[0057] U={u0,...,u i ,...,u n-1} (1)

[0058] Where i = 0, 1, ..., n-1, and n is the number of samples of the aircraft sensor;

[0059] The transmission delay T of the sampling value from the sensor signal acquisition to the software filtering processing delay is a known quantity. Specifically, the transmission delay T delay It can be that the transmission delay of each sampling value from the sensor acquisition signal to the software filtering processing is a constant, that is:

[0060] T send -T recieve＝ T delay＝ Constant (2)

[0061] Alternatively, the transmission delay T delay is a variable, but can be measured in other ways, namely:

[0062] T send -T recieve＝ Tdelay＝ Measurable values (3)

[0063] Where: T send Indicates the time when the sensor sampling value is sent;

[0064] T recieve Indicates the time at which the filter processes the sample value.

[0065] Preferably, before establishing the interpolation function of each channel of the sensor sampling value, the difference quotient calculation is first required. The interpolation quotient expression of each order of the sampling value is:

[0066] First-order difference quotient expression:

[0067]

[0068] Second-order difference quotient expression:

[0069]

[0070] Therefore, according to the recursive formula, the N-order difference quotient expression can be obtained:

[0071]

[0072] After calculating the difference quotients of the sampling values, the interpolation function P can be established. n (u) The expression is as follows:

[0073] P n (u)=f(u0)+f[u0,u1](u-u0)+f[u0,u1,u2](u-u0)(u-u1)+…+f[u0,u1,…,u n ](u–u0)(u-u1)…(uu n-1 ) (8)

[0074] Wherein, u represents the sampling correction value at the current moment.

[0075] According to formula (1)-formula (8) and the real-time sampling values of each sensor and the corresponding sampling time, when the three continuous discrete moments t0, t1, t2 and the corresponding moment sensor sampling values [t0, u(t0)], [t1, u(t1)], [t2, u(t2)] are obtained, the interpolation function P is used n (u) can obtain the correction result of the sampling value u(t) at a certain time t:

[0076]

[0077] It can be seen that in the case of equal sampling interval time, let T be the sampling interval, the above formula can be simplified to:

[0078]

[0079] From the numerical analysis, we know that the interpolation function error is R2(t), which is expressed as follows:

[0080]

[0081] Wherein, u′″(ξ) is the third-order derivative of the sample value u(t) at time t at t=ξ, ξ∈[t0,t2].

[0082] Preferably, for each measured sample value u i (t i ), set d adjacent measurement points as a filtering reference interval, and the filtering reference interval is shifted by m measurement points over time to update the iterative filtering sequence; then, using formula (10), iterate the corrected sample value queue and enter the final output filtering queue. Furthermore, d is generally 3-5.

[0083] like Figure 2 As shown, for each measured sample value u i (t i ), set 5 adjacent measurement points as a filtering reference interval. During the analysis and processing, the filtering reference interval moves m measurement points one by one over time, and analyzes and compares each sampling value before filtering.

[0084] like Figure 2 As shown, l = 6, m = 3. If the permissible trust threshold increment between adjacent sampling values is known to be Δ (usually obtainable from the physical properties of the sampling values), then the method for eliminating the p-fold error of each sampling value is:

[0085] if |u i (t i )-u i-p (t i-p )|≥p×Δ, then u i (t i ) is corrected using the following formula.

[0086]

[0087] Where p∈(1,…,l), j is greater than The smallest integer, u' i (t i ) is the corrected result.

[0088] The present invention realizes the trend estimation of aircraft fuel quantity measurement data, makes it closer to the real state, eliminates the error of oil level fluctuation, further improves the measurement accuracy of fuel, and has more superior performance.

[0089] Example 2:

[0090] A method for synchronously collecting aircraft sensor data based on interpolation is applied to aircraft fuel quantity sensors, such as Figure 1 As shown, the following steps are included:

[0091] Step 1: Under the premise of the same sampling frequency, the fuel sensors of each aircraft perform asynchronous sampling;

[0092] Step 2: Set the filter queue address, confidence interval length, and filter limit Δ limit And the filter queue upper limit Δ upper , filter queue lower limit Δ lower Specifically, the filter queue address defines the storage starting location of the sample value, which is convenient for finding the initial value of the filter; setting the confidence interval length, that is, defining the response speed and interval length of the filter, is convenient for finding the number of data that need to be filtered; the filter limit defines the value range of the sample value to eliminate obviously false data; the filter queue upper limit Δ upper and filter queue lower limit Δ lower The confidence interval range of the filter sequence is defined to distinguish the credibility of the sampling values; secondly, the output interface of the credible sampling values is defined.

[0093] Step 3: Receive the aircraft fuel sensor sample value, then first use the filter limit Δ limit Remove obvious false data; then use the upper limit Δ of the filter queue upper and the lower limit Δ lower To distinguish the credibility of the sampled values, we can get the credible sampled value U = {u0,...,u i ,...,u n-1}; Finally, the reliable sampling value is output. The filtering limit Δ limit It is the valid range of oil sampling value, which is a positive value.

[0094] Since the actual fuel consumption of the aircraft is reduced at a relatively stable fuel consumption rate, the concept of reliable fuel sampling value is introduced here: reliable sampling value u i Changes within a certain range. When it changes downward, the credible sampling value u i The minimum relative base oil volume is not less than Δ lower , can also change upward, but the maximum cannot exceed Δ upper .

[0095] After the filter starts running, the sample value S i , it is necessary to fill the data queue after n measurement cycles as the basis for subsequent comparison. On the basis of the queue being filled, first S i and filtering limit Δ limit By comparison, obvious false data can be rejected; sorting by the original order, if Si Less than the filtering limit Δ limit And in [Δ lower , Δ upper ] interval, it is considered to be credible and enters the confidence interval.

[0096] Step 4: If n cycles have passed, the reliable sampling value u i Enter the queue, otherwise go to step 3; if the sensor sampling value falls within the confidence interval condition, use formula (1) to obtain the initial sequence of sampling values;

[0097] Step 5: Determine the transmission delay based on the time from the sensor collecting the signal to the filter processing according to formulas (2) and (3);

[0098] Step 6: According to formula (4)-formula (8), after calculating the difference quotients of each order of the sampling value, the improved interpolation function P is established. n (u);

[0099] Step 7: Use the interpolation function P n (u) can obtain the expression (9) of the corrected result of the sampling value u(t) at a certain time t. According to the input sensor sampling value and the assumption that the sampling interval time is equal, the formula (9) can be used to simplify the expression (10) of the result after the synchronization of the aircraft sensor sampling value data based on the improved interpolation method.

[0100] Step 8: Finally, sort the filter according to its characteristics, such as Figure 2 As shown, for each measured sample value u i (t i ), set d adjacent measurement points as a filtering reference interval, and the filtering reference interval moves m measurement points one by one over time. According to formula (12), the sampling value unit is moved in sequence to update the filtering queue, and then the revised result of the sampling value queue is iterated by repeatedly using formula (10) to enter the final output filtering queue.

[0101] The present invention updates the iterative Newton interpolation results by detecting errors, allowing for credible threshold increments, and moving the filter reference interval. The Newton interpolation method used in this invention is the first to be implemented in the field of fuel measurement, filling the gap in the fuel quantity filtering algorithm. It also moves the filter reference interval and updates the results of the iterative filtering queue by analyzing the sampling value error and allowing for credible threshold increments to be added. The present invention expands the application field of the Newton interpolation algorithm and corrects the error compensation results of the algorithm. It is the first innovative verification and has industrial application value and practical application background. It sets a benchmark and solution for the industry and has significant application space for promotion in the field of aviation measurement.

[0102] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Any simple modification or equivalent change made to the above embodiment based on the technical essence of the present invention shall fall within the scope of protection of the present invention.

Claims

1. A method for synchronously collecting aircraft sensor data based on interpolation method, characterized in that: The following steps are involved: Step S1: Under the premise of the same sampling frequency, each aircraft sensor performs asynchronous sampling; Step S2: Receive each sampling value and filter to obtain the credible sampling value U={u0,...,u i ,...,u n-1 }, obtain the initial sequence of sampling values; where i = 0, 1, ..., n-1, and n is the number of samples of the aircraft sensor; then, calculate the sampling time of each sampling value based on the transmission delay; Step S3: using the sampling value and the sampling time as two columns of original data, an interpolation function of each sampling value is obtained based on an interpolation algorithm; Step S4: Select a fixed time sequence, calculate the correction value corresponding to the sampling value by the interpolation function of step S3, and use it as the new sampling value. The sampling time is fixed in the same sequence to achieve data synchronization of the sampling value.

2. The method for synchronously collecting aircraft sensor data based on interpolation method according to claim 1, characterized in that: The step S2 comprises the following steps: Step S21: Set the filter queue address, confidence interval length, and filter limit Δ limit And the upper limit of the filter queue Δ upper and the lower limit Δ lower ; Step S22: Using the filter limit Δ limit Keep out obviously false data; Step S23: Using the upper limit Δ of the filter queue upper and the lower limit Δ lower Filter the sample values and get the values that fall within [Δ lower , Δ upper ] interval of credible sampling value; Step S24: Divide the original sampling values detected by the aircraft sensor into n units according to the original order to determine the initial sampling value sequence.

3. The method for synchronously collecting aircraft sensor data based on interpolation method according to claim 1 or 2, characterized in that: In step S2, a transmission delay is determined based on the time from when the sampled value is collected by the sensor to when the sampled value is processed by the filter; the transmission delay is a constant or a measurable variable.

4. The method for synchronously collecting aircraft sensor data based on interpolation method according to claim 1, characterized in that: The step S3 comprises the following steps: Step S31: First, calculate the difference quotient. The interpolation quotient of each order of the sample value is expressed as: First-order difference quotient expression: Second-order difference quotient expression: By analogy, the expression of the N-order difference quotient is: Step S32: Then, establish the interpolation function P of each sampling value n (u) is: Where u is the sampling correction value at the current moment.

5. The method for synchronously collecting aircraft sensor data based on interpolation method according to claim 4, characterized in that: In step S4, when obtaining three consecutive discrete moments t0, t1, t2 and the sensor sampling values at the corresponding moments [t0, u(t0)], [t1, u(t1)], [t2, u(t2)], the interpolation function P is used. n (u) The correction result of the sampling value u(t) at time t is: In the case of equal sampling intervals, formula (9) is simplified to: Where T is the sampling interval.

6. The method for synchronously collecting aircraft sensor data based on interpolation method according to claim 5, characterized in that: The interpolation function error R2(t) is: Wherein, u″′(ξ) is the third-order derivative of the sampling value u(t) at time t at t=ξ, ξ∈[t0,t2].

7. The method for synchronously collecting aircraft sensor data based on interpolation method according to claim 5, characterized in that: In step S4, for each measured sample value u i (t i ), set d adjacent measurement points as a filtering reference interval, and the filtering reference interval moves m measurement points one by one over time to update the iterative filtering sequence; then use formula (10) to iterate the result of the corrected sampling value queue and enter the final output filtering queue.

8. The method for synchronously collecting aircraft sensor data based on interpolation method according to claim 7, characterized in that: If the permissible credible threshold increment between adjacent sample values is known to be Δ, then the method for eliminating the p-fold error of each sample value is: if |u i (t i )-u i-p (t i-p )|≥p×Δ, then u i (t i ) is corrected using the following formula: Where: j is greater than The smallest integer; u' i (t i ) is the corrected result.

9. The method for synchronously collecting aircraft sensor data based on interpolation method according to any one of claims 1 to 8, characterized in that: Used in aircraft fuel level sensors.

Images